The behind-the-scene efficiency of Hong Kong’s hospitals and MTRC subway system is being transformed by artificial intelligence (AI) rostering and scheduling software developed by researchers at City University of Hong Kong.
Until recently, rostering of the Hospital Authority’s more than 48,000 staff at 44 hospitals was handled manually. Now, a project to computerise and optimise the task is being rolled out with a resulting increase in staff morale and improved service for patients.
At the MTRC a similar, but much larger, system is currently being verified to ensure maintenance staff with the required skills, plus equipment and material, are in place at the right time for trains and track on all six subway lines to be well and timely maintained.
PI of the projects, Dr Andy Hon-wai Chun, said of the MTRC system: “Our main task is to make sure all the jobs can be done safely. In some places, electricity needs to be off, in other places it needs to be on.
“As well as ensuring the whole operation is safe, we also ensure that adequate resources are available, such as train drivers, supervisors, and equipment.”
He added: “In the Hospital Authority’s case, there are many considerations like rostering equal work loads for nurses; scheduling them to work 44 hours a week, including so many night shifts and, of course, a night shift can’t then be followed immediately by a day shift. There are also requests for days off, contract considerations, skill levels, and many other inputs.”
The tailored computer programs devised by Dr Chun and his researchers combine constraint programming techniques with AI algorithms with the result that complex problems can be solved in a very short time.

Principal Investigator
Dr Andy Hon-wai Chun : andy.chun@cityu.edu.hk